AWS announced last week the addition of new functions to your Lambda platform. The new features introduced by AWS Lambda include support for the AVX2 instruction set, support for container images.
AWS Lambda now can provide functions with storage up to 10GB memory and 6 vCPUs (virtual processors), which will allow developers to create more compute-intensive functions to obtain the resources they need.
For those unfamiliar with AWS Lambda, you should know that is an event-driven serverless platform provided by Amazon as part of your Amazon Web Services cloud offering. Serverless computing does not mean there is no server. Esto means developers no longer have to worry about computing needs, storage and memory, because the cloud provider, AWS in this case, takes care of it.
This allows developers to code the application instead of implementing resources. The goal of AWS Lambda, compared to AWS EC2 (Elastic Compute Cloud), is to make it easier to build smaller on-demand applications that respond to events and new information.
AWS Lambda supports the safe execution of native Linux executables using a supported runtime, such as Node.js. For example, Haskell code can be run on Lambda.
Container image support makes it easy to to business users the use of a consistent set of tools for security scanning, code signing, etc. Also allows the maximum code pack size for a function to be increased to 10 GB.
This feature blurs the line between Lambda and containers and can be confusing, so it's safe to start by understanding what this functionality is and is not. Therefore, note that this feature is not a replacement for AWS ECS (Amazon Elastic Container Service) or AWS Fargate.
You can't run long-lived services in Lambda, your code is always bound by Lambda's invocation pattern (that is, it only runs when the function is called). Function calls are always linked for the same maximum duration of 15 minutes.
In addition, the container image must interact with the Lambda Runtime API to request events and send responses, as well as a custom Lambda runtime. This new feature allows you to send the content of a Lambda function as a container image instead of a zip file.
It also runs the base image as is, so you can use a Linux image, such as Alpine or Debian, In addition, an arbitrary base image can be used with which the open source AWS Lambda Runtime Interface (RIC) client can be used to make your base image compatible with the Lambda Runtime API.
Container images up to 10GB can now be packaged, which is significantly higher than the 250MB limit on deployment package size. Like a custom Lambda Runtime, the container image must have a bootstrap file that interacts with the Lambda Runtime API to request events and send responses.
Starting today, you can allocate up to 10GB of memory for a Lambda function. This represents an increase of more than three times over the previous limits. The Lambda function allocates CPU and other resources linearly, proportional to the amount of memory configured. This means that you can now have access to up to 6 vCPUs in each runtime environment, ”the company wrote in a blog post announcing the new capabilities of AWS Lambda.
This allows you to specify the location of the boot file using the "ENTRYPOINT" and "CMD" parameters in the Docker file.
You can also configure the working directory using the "WORKDIR" parameters and configure the environment variables with the "ENV" parameter. After you have created the Docker image, you must deploy the image to the Amazon Elastic Container Registry (ECR). Additionally, the Lambda service must be granted the necessary Identity and Access Management (IAM) permissions to access the repository and get the container image.
Source: https://aws.amazon.com/blogs